An Evaluation of Confirmatory Factor Analysis of Ryff’s Psychological Well-being Scale in a Persian Sample

This paper examines the construct validity and reliability of the Psychological Well-Being Scale (PWBS) according to the Persian culture and language. Participants (N=577) from a population study of university students were chosen. Confirmatory factor analysis using AMOS software was performed in two steps. In step one, 18 models derived from 3-, 9-, and 14-item forms that emphasized gender differences in addition to first and second order constructs were compared. In step two, the 9-and 14-item forms were compared aiming modification. In step one only the 3-item form achieved reasonable indices. Allowing for gender differences did not result in a model fit in the 9-and 14-item forms. To achieve a model fit with additional items, in step two, models that used the 9and 14-item forms with a second order factor structure regardless of gender differences was performed for modification. This modification allowed for greater potential for comparison with other models in order to achieve good indices. The results in step two indicated that after deleting of some items from the two models, the 14-item model showed better construct validity and reliability compared to the model based on the 9-item form in the Persian culture.